In the conventional pyrolysis of biomass, typically fast pyrolysis carried out in an inert atmosphere, a dense, acidic, reactive liquid bio-oil product is obtained, which contains water, oils, and char formed during the process. Much of the oxygen present in biomass ultimately resides in the bio-oil, thereby increasing its chemical reactivity. Characteristic total acid numbers (TAN) of conventional bio-oil are high, in the range of 100-200, indicating it is a highly corrosive material. Moreover, this product tends to undergo polymerization, is generally incompatible with petroleum hydrocarbons due to water miscibility and very high oxygen content (on the order of about 40% by weight), and has a low heating value. The unstable bio-oils of conventional pyrolysis tend to thicken over time and can also react to a point where hydrophilic and hydrophobic phases form. As a result, transportation and utilization of this product are problematic. Also, it is difficult to upgrade this product to a liquid hydrocarbon fuel, due to the retrograde reactions that typically occur in conventional pyrolysis processes, including fast pyrolysis. Dilution with methanol or other alcohols has been shown to reduce the activity and viscosity of the formed bio-oils, but this approach is not considered practical or economically viable, due to the large amounts of unrecoverable alcohol that are required to stabilize pyrolysis liquids.
The removal of char, generated by conventional pyrolysis, from the liquid pyrolysis product while it is still in the vapor phase presents an additional technical challenge. Significant amounts of oxygen and free radicals in the pyrolysis vapors remain highly reactive and form a pitch-like material upon contact with char particles on the surface of a filter or other solid separator. Consequently, devices used to separate char from the hot pyrolysis vapors can become quickly plugged, due to the reactions of char and pyrolysis vapor constituents that occur on and within the layer of char on the surfaces of such devices, as well as within the pores of porous filter elements. Finally, it is noted that the upgrading of pyrolysis oils, using conventional hydroconversion processes, consumes large quantities of H2, and extreme process conditions, including high hydrogen pressures needed to meet product quality requirements, make such processes uneconomical. The reactions are inherently out of balance in that, due to the high pressures required, too much water is created while too much H2 is consumed. In addition, conventional hydroconversion reactors can rapidly develop high pressure differentials, due to reactive coke precursors present in the pyrolysis oils or from coke produced as a result of catalysis.
More recently, the use of hydrogen in biomass pyrolysis (i.e., hydropyrolysis) has been disclosed. For example, hydropyrolysis processes taught in U.S. Pat. No. 8,492,600 have been found to overcome a number of the drawbacks of conventional fast pyrolysis processes, including those described above, and have led to a number of other processing advantages. Despite these and other improvements, the use of certain feedstocks, such as lignocellulosic biomass, in hydropyrolysis results in a defined distribution of hydrocarbons, following deoxygenation, that may be predominantly suitable for certain applications (e.g., in the production of an aromatic-rich, renewable gasoline fraction or blending component) but not others (e.g., in the production of a high cetane, renewable diesel fraction or blending component). The art of hydropyrolysis, however, would benefit significantly from processing options that allow flexibility in terms of providing hydrocarbon-containing product fractions in varying yields and compositions, and with desired product quality attributes (e.g., cold flow properties in the case of a diesel boiling range fraction or octane number in the case of a gasoline boiling range fraction). Such processing options could be adapted as needed to meet changing end-product demands in the art of renewable fuels, thereby maximizing overall value and process economics.